Labeling of samples is a recurrent and time-consuming task in data analysis and machine learning and yet generally overlooked in terms of visual analytics approaches to improve the process. As the number of tailored applications of learning models increases, it is crucial that more effective approaches to labeling are developed. In this paper, we report the development of a methodology and a framework to support labeling, with an application case as background. The methodology performs visual active learning and label propagation with 2D embeddings as layouts to achieve faster and interactive labeling of samples. The framework is realized through SoundscapeX, a tool to support labeling in soundscape ecology data. We have applied the framework to a set of audio recordings collected for a Long Term Ecological Research Project in the Cantareira-Mantiqueira Corridor (LTER CCM), localized in the transition between northeastern São Paulo state and southern Minas Gerais state in Brazil. We employed a pre-label data set of groups of animals to test the efficacy of the approach. The results showed the best accuracy at $94.58\%$ in the prediction of labeling for birds and insects; and $91.09\%$ for the prediction of the sound event as frogs and insects.

Feature Analysis has become a very critical task in data analysis and visualization. Graph structures are very flexible in terms of representation and may encode important information on features but are challenging in regards to layout being adequate for analysis tasks. In this study, we propose and develop similarity-based graph layouts with the purpose of locating relevant patterns in sets of features, thus supporting feature analysis and selection. We apply a tree layout in the first step of the strategy, to accomplish node placement and overview based on feature similarity. By drawing the remainder of the graph edges on demand, further grouping and relationships among features are revealed. We evaluate those groups and relationships in terms of their effectiveness in exploring feature sets for data analysis. Correlation of features with a target categorical attribute and feature ranking are added to support the task. Multidimensional projections are employed to plot the dataset based on selected attributes to reveal the effectiveness of the feature set. Our results have shown that the tree-graph layout framework allows for a number of observations that are very important in user-centric feature selection, and not easy to observe by any other available tool. They provide a way of finding relevant and irrelevant features, spurious sets of noisy features, groups of similar features, and opposite features, all of which are essential tasks in different scenarios of data analysis. Case studies in application areas centered on documents, images and sound data demonstrate the ability of the framework to quickly reach a satisfactory compact representation from a larger feature set.

Para a análise de Paisagens Acústicas os especialistas precisam conhecer os eventos em um determinado ambiente. Uma das formas de se adquirir esse conhecimento é através do som, isto é, de paisagens acústicas....

The importance of the classification of forest fires through computational methods lies in their applicability for early detection and timely warning to prevent their spread. According to the state of the art, the classification of forest fires was approached by proposals based on machine learning (ML) and deep learning (DL), more specifically on Convolutional Neural Networks (CNN). In practice, DL-based methods prove to be superior to traditional ML methods. Recently, FFireNet, a CNN model that presents optimal results for classifying images of forest fires, was proposed. On this basis, in thiswork, we present a methodology to create CNN models by transfer learning applied specifically to the classification of forest fires, where techniques such as data augmentation, transfer learning, transfer learning and fine tuning were included. The proposal allowed us to define E-FFireNet, an efficient model with superior results than FFireNet. The experiments showed classification results of 99,73 (accuracy), 99,48 (precision), 100,0 (recall) and 0,263 (error).

O consumo de energia elétrica é uma preocupação global devido ao seu uso de forma ineficiente e descontrolado, resultando em gastos excessivos. Nesse contexto, o presente artigo propõe a aplicação de técnicas de inteligência artificial, em particular o machine learning, para analisar o consumo de energia residencial. Foram utilizados dados obtidos por meio de sensores que passaram por um pré-processamento para eliminar instâncias com dados incompletos. Uma categorização do consumo de energia é apresentada, permitindo que a inteligência artificial implementada realize a classificação correspondente. Também foi realizada uma análise de técnicas para dados desbalanceados referente ao consumo de energia, assim, de acordo com a quantidade de experimentos executados, são fornecidas recomendações para balanceamento com a menor perda de informação possível. A validação da metodologia foi feita comparando os rótulos verdadeiros com as categorias preditas. Os resultados demonstraram a eficácia da proposta na classificação, alcançando 98% de acurácia. Além disso, as visualizações por meio de UMAP, t-SNE e Coordenadas Paralelas destacaram a separação das instâncias de distintas categorias, indicando a eficácia da metodologia.